An alternative approach for direct APC using scatterometry data

Scatterometry is a promising method, capable of providing accurate profile information for a large range of applications. However, applying scatterometry to the production environment and applying it to APC is still difficult. In this paper we propose an alternative approach in which we apply a Neural Network to directly correlate scatterometry raw data and the lithography process control parameters. The proposed method is much easier to use than normal scatterometry, and can therefore be applied to APC much faster.